New Jersey Water Science Center
LONG ISLAND-NEW JERSEY NAWQA
USGS IN YOUR STATE
USGS Water Science Centers are located in each state.
Long Island-New Jersey (LINJ) Coastal Drainages Study
FY1998 Workplan—Occurrence and Movement of Contaminants through the Urban Hydrologic CycleFebruary 1, 1997
Emissions from vehicles and industry are non-point sources from a hydrologic cycle perspective because of the dispersive effect of atmospheric transport. Because of water/air phase partitioning, precipitation and direct runoff introduce contaminants directly to streams, rivers, and lakes. Point sources to surface water include effluent discharge from industrial and municipal treatment plants. Precipitation falling on land and moving through the unsaturated zone (recharge) combined with the process of diffusion of contaminants from the atmosphere through the unsaturated zone is a potential low-level aerially extensive source of contaminants to shallow groundwater. This non-point source, referred to as atmospheric deposition, has been proposed as an explanation for the frequent detection of the fuel oxygenate MTBE at low-level concentrations in shallow groundwater in urban areas. Urban runoff, storm sewers, and detention basins are other potential non-point sources of contaminants to shallow groundwater. Leaking underground storage tanks and other spills are point sources of BTEX, fuel oxygenates (eg. MTBE), chlorinated hydrocarbons, and other chemicals. Once introduced into groundwater, contaminants can move along flow paths to deeper parts of the aquifer system eventually discharging to wells or surface water bodies. In all compartments of the hydrologic system contaminants may biodegrade to other chemicals of concern or less harmful molecules.
In the urban environment, concern is not restricted to VOC's such as BTEX and MTBE as other compounds may be of equal or greater concern. Pesticide and nutrient usage is widespread because of weed control, lawn/garden care, and agriculture mixed in the metropolitan landscape. New housing frequently occupies former farmland and inherits the chemical usage history. At the basin scale, this chemical usage is a potential non-point source.
The table below itemizes components of a comprehensive urban study and summarizes the time scale for implementation for the Glassboro Region. This list is a result of collaboration between LINJ, the VOC Synthesis team, and Lehn Franke. The list is comprehensive to provide a national perspective, however, the relative importance of each item is dependent upon geography and water supply issues of a particular study area. In this regard it is important to note that the water supply concerns for the Glassboro region are groundwater dominated. Coordination with the core LINJ-NAWQA project was necessary from the beginning to derive synergy, cost effectiveness and holism. In the case of LINJ; land use survey, flow path study, and stream sampling have been coordinated.
TABLE 1-- Components of a Comprehensive Urban Study
Selection of study area (FY95)
The New Jersey Department of Environmental Protection (DEP) has recommended increased development of the surficial Kirkwood- Cohansey aquifer system within the Glassboro region to meet a portion of the water demand caused by the explosive suburban growth. The Glassboro region is within DEP Water Supply Critical Area #2, which is characterized by a large cone of depression in the deeper, confined Potomac-Raritan-Magothy aquifer system caused by heavy pumpage in Camden County, northwest of the Glassboro Region. Recent mandates place constraints on additional development of the confined aquifer system within the Critical Area. The Tri-County pipeline will import water from the Delaware River and provide adequate supply to communities within Critical Area #2 north and west of Glassboro, however, additional water supply from the Kirkwood-Cohansey is needed for communities that will not have access to the pipeline.
VOC-atmospheric-synoptic sampling was conducted across the study area from June 6 - September 30 1996 to determine spatial variability and select locations for fixed sites. Based on this study three fixed sites were selected. One fixed site is in Turnersville at a location of high traffic, another is at Rowan College in Glassboro at a location of average traffic and the third is at Coles Farm in Pole Tavern at a location of low traffic (figure 2). In October 1996, sampling on a 13-day interval began at each of the three fixed locations. As of December 1996 MTBE and BTEX have been detected in all samples.
Through FY97, VOC 24-hour composite samples will be collected at the three locations on a 13-day cycle. A field blank consisting of a cartridge (set in the field and not pumped) will be included every 13-days rotating the blank among the three sites. Temperature, atmospheric pressure, relative humidity, and general weather condition will be obtained from nearby weather stations. SCL will be used throughout FY97. When OGI is ready to process samples (estimate-April, 97) then we will simultaneously collect samples at the three sites on the 13-day schedule for analysis for OGI, SCL, and NJDL throughout FY97. OGI has recommended a quality control strategy involving travel blanks, field blanks, and spiked cartridges which will be implemented. Simultaneous data collection in FY97 for both OGI and SCL will provide an extensive field test of two GC/MS methods and provide a refinement of the method ultimately recommended to NWQL. The purpose of the NJDL data set is to test a simpler, less costly (The N.J. district purchased a thermal desorption unit in FY96) GC/FID method. The NJDL list of analytes will be expanded as indicated from the more comprehensive analysis provided by SCL or OGI. It is anticipated that the data base as well as unsaturated zone gas data can be maintained for long-term study or study of other areas in N.J. in a cost effective manner using NJDL with occasional MS verification from NWQL or another lab. Atmospheric sampling will continue through FY98, however, at this time the service lab is uncertain because the time frame of NWQL incorporation of VOC air samples can not be determined.
VOC-precipitation sampling will be initiated by March 1997. The purpose of this sampling is to determine if precipitation and atmospheric gas is in equilibrium with respect to select VOC. If equilibrium is an appropriate approximation then atmospheric loading of VOC can be inferred from an air quality record, precipitation record, and Henry's Law coefficients, thereby eliminating the need to collect precipitation-water quality measurements in future studies. This work will be conducted at the Sagamore Junior High School in Suffolk County, Long Island. This location was selected to take advantage of frequent VOC- atmosphere monitoring initiated by SCL in April, 1996 which will facilitate equilibrium hypothesis testing. The atmosphere is sampled every 6 days (24-hour composite samples), in addition, precipitation and other parameters are collected. Precipitation will be sampled simultaneously with the atmosphere for 10-15 precipitation events in FY97. Precipitation will be collected and then sent to NWQL for analysis. The method of collecting a VOC precipitation sample has not yet been determined and several methods are under consideration. A sample of the atmosphere will be collected during precipitation with the sampling pump on continuously at the maximum allowable flow rate until the required sample volume is obtained (a period on the order of several hours). SCL has experienced difficulty in analyzing cartridges set out on days of precipitation, therefore the 6-day record could be of greater importance in supporting the study if air samples taken during storms are problematic. If the equilibrium studies are problematic then VOC precipitation sampling will be initiated at Rowan College in late FY97 through FY98.
A nest of unsaturated zone gas sampling probes has been installed at both Rowan College and Sagamore School. These nests will be sampled quarterly as described below in the unsaturated zone section. In addition to the quarterly sampling, a shallow probe exists in the nest at each location at about 3 feet below land surface. The shallow probes will be sampled about every month at both sites to provide a time series with less deviation to compare with the atmospheric record.
Pesticides in atmosphere and precipitation
One location has been selected for atmospheric and rainwater sampling for pesticides in FY97 - Bunker Hill Church in Washington Twp., N.J. At Bunker Hill Church sampling began in December 1997. Weekly composite samples for both PUF and GFF will be collected every other week through the end of March, 1997. From April through August, 1997 weekly composite samples will be collected to correspond with the heavy usage season. Precipitation samples will be collected on the same schedule and analyzed for pesticides, nutrients, major ions and ph. Unfortunately VOC sampling of the precipitation is inappropriate because of volatile losses that would occur over the sample integration period. Field blanks (PUF-non particulate only) will be collected once a month. The total sampling commitment is about 30 for each of GFF and PUF, and 20 precipitation analysis. The program for FY98 will be decided based on this data and collaboration with other researchers.
Gas-phase sampling for VOC and major gases
Installed unsaturated-zone instrumentation is of two types: probe nests and single probes associated with shallow groundwater observation wells (see groundwater section). A probe consists of a buried length of 1/4-inch stainless-steel tubing with a 4 inch teflon tube screen. In nests the probes are separated from one another with bentonite to allow for collection of a gas sample from various elevations in the unsaturated zone. Probe nests provide data for analyzing the occurrence and movement of MTBE and other VOCs through the unsaturated zone. Samples of MTBE, BTEX, and other VOCs are collected by placing a multisorbent cartridge in line with a peristaltic pump. The multisorbent cartridge is of the same type used to collect atmospheric samples and, therefore, the same analytical schedule is used. Whole air samples taken with a syringe allow for the determination of major gases (oxygen, nitrogen, carbon dioxide) with GC/TCD at the NJDL.
Gas sampling probe nests are located at Turnersville (installed 6/96), Rowan College (installed 9/96), and Sagamore Junior H.S., Long Island (installed 12/96). At these locations VOC atmospheric sampling is also being conducted. Single gas sampling probes were installed during summer-fall of 1996 at all 72 observation wells of the extended land use survey of the Glassboro region. MTBE and BTEX have been detected in the unsaturated zone at the Turnersville site at concentrations consistent with atmospheric concentrations. Experimental sampling done at Turnersville has provided guidance on sample volume size for unsaturated zone gas sampling.
In FY97 the probe nests at Turnersville, Sagamore Junior High School, and Rowan College will be sampled quarterly (March, June, September, December) for VOC and major gases. These three locations provide data to study the movement of VOC due to atmospheric deposition because of the co-existence of atmospheric sampling. The need for additional nests to study atmospheric deposition will be determined based on data from these locations.
In FY97 each single probe associated with a shallow groundwater observation well that has a VOC hit above a concentration level to be determined will be sampled. The purpose of this data is to support source delineation. The scope of this sampling is not yet defined as we await data from NWQL for the 72 well survey just completed. Approximately 5 additional probe nests will be installed at select locations of VOC hits to help determine whether or not the source originated at land surface. These additional locations are the same as those discussed in the UNSATURATED ZONE/SHALLOW GROUND WATER interface section below. The nests will be sampled for VOC (SCL and NJDL) and major gases (NJDL).
Water and total phase sampling for pesticides, nitrate, and other low volatility compounds
If a gaseous phase signature can be measured, then equilibrium calculations are assumed to give satisfactory estimates of the total occurrence of the chemical in the unsaturated zone. Many chemicals of interest (nitrate and pesticides for example), however, do not significantly partition in the gaseous phase, therefore, unsaturated zone occurrence monitoring needs to be expanded to include water and total phase sampling. Lysmeters will be used to obtain water samples and core taken for total sediment. The focus will be at well locations demonstrating elevated nitrate and pesticide concentrations. Sampling will be initiated in FY98, however we will install lysymeters in FY97 to let them equilibrate.
Transport modeling parameters
Transport modeling for the unsaturated zone is required to obtain rates of movement and loading at the water table from occurrence data. It provides a dynamic link between land-surface sources and shallow groundwater. This section focuses on data collection to support transport modeling. The application of unsaturated zone transport modeling to help the NAWQA program interpret land use surveys is discussed further in the section below-unsaturated zone/shallow groundwater interface (with Franke).
During the installation of the 72 shallow observation wells in 1996, continuous core from land surface to the water table was taken at 60 of the 72 sites. These cores allow for determination of unsaturated zone lithology, recharge, effective diffusion coefficients, moisture content, porosity, bulk density, organic content, and grain size distribution. These parameters are required to support transport modeling and provide area-wide mapping of a vulnerability index for the aquifer. Moisture content was measured for each 2-foot section within days of collection, then the core was stored. In FY97 laboratory experiments and parameter measurements will be conducted on a subset of core sections selected to represent the range of unsaturated zone sediments encountered.
Extended land-use survey
The shallow-groundwater monitoring network, installed during summer-fall 1996, was designed to randomly sample recently recharged groundwater across the Glassboro region. It consists of 72 observation wells across the study area (figure 2). The network constitutes an extended land use survey as well locations are divided among major land-use categories, with 30 wells located in newly developed (<25 years) urban land, 14 wells located in older urban land, 15 wells located in agricultural land, and 13 wells located in undeveloped land. A grid-based random-sampling approach was used to select well locations within each land-use category to ensure unbiased data collection. The wells are of PVC construction and typically are screened over a 2-foot interval set about 10 feet below the water table level at the time of installation. Sampling of all 72 wells for the full suite of NAWQA analytes (VOC, pesticides, and major inorganic constituents) began in fall 1996 and was completed in December 1996. A summary of the data available from NWQL at this time is provided in Table 2. In addition 12 wells were sampled in December 1996 for dissolved CFC's. The 12 unsaturated zone-gas probes corresponding to the wells were also sampled in January for CFC and VOC (see vertical flow path sampling section below).
TABLE 2-- Water quality data from 72 wells in the land use survey conducted
56 wells with VOC data - (14 ag, 11 und, 22 new urb, 9 old urb.)
Flowpath study and transport modelingA preliminary three-dimensional groundwater-flow model of the Kirkwood-Cohansey aquifer in the Glassboro region was developed in 1996. The model provides estimates of the time required for water to move along any flowpath from the water table to eventual discharge at a stream or a water-supply well. This flowpath analysis will be used to evaluate existing wells and future well locations for inclusion into a monitoring network that will allow for sampling of groundwater of different ages at greater depth to supplement the shallow-groundwater sampling network. Chemical transport modeling to determine the relevance of compound occurrence to the water supply derived from the surficial Kirkwood-Cohansey aquifer system can be proposed for FY98.
To date the entire groundwater part of the workplan has been conducted within the core LINJ project and interpreted as a combination land use survey/flow path study. Flow path study and transport modeling is supported through 1998 by the core project. This data collection and modeling activity is also a fundamental part of the comprehensive urban study. Because of the large expense of well selection, drilling, and sampling it has been necessary to coordinate the comprehensive urban study with the core LINJ project from the earliest stages.
Vertical flowpath samplingFifteen of the 72 single unsaturated-zone-gas probes were sampled in December 1996 for VOC and CFC to test methods of CFC analysis and to provide unsaturated zone CFC data to compare with CFC groundwater data from the underlying observation wells (which were also sampled in December 1996 for CFC). The fifteen locations provide representation of all land use categories and span the range of unsaturated zone thickness. This data will determine if age differences of water across the water table over a scale of 10-15 feet are decernable with CFC analysis. The relevance of this 15 location synoptic with respect to VOC and pesticide hits is uncertain because at the time of sampling water quality data from the extended land use survey was not available.
The 72 land use survey wells installed in summer-fall 1996 are of PVC construction and typically are screened over a 2-foot interval about 10 feet below the water table level at the time of installation. The added value of these wells, in regard to this component of the workplan, lies in their accoutrements which consist of, in addition to the standard observation well, a stainless steel gas sampling probe (see unsaturated zone section above) located about 3 feet above the water table and a groundwater sampling probe located about 3 feet below the water table. The shallow groundwater probe consists of 1/4-inch plastic tubing strapped to the well casing which allows for withdrawal of a groundwater sample with a peristaltic pump, provided the water table is no deeper than 25 feet. This instrumentation allows for sampling along a vertical flow path across the water table for VOC. For VOC, pesticides, and nitrate it provides the opportunity to relate water-quality data as a function of depth below the water table, a capability not yet available to NAWQA. At 10 feet below the water table flow could be horizontal.
In FY97 we will select 5 well locations from the 72 well network that had MTBE concentrations greater than .2 ug/L (and perhaps other VOC) and 5 well locations for which there were higher-level pesticide and nitrate concentrations (the target pesticide has not yet selected). For MTBE/VOC locations a gas sampling probe nest will be installed several months before sampling which will occur early in FY98. These nests are the same as the ones discussed above in the UNSATURATED ZONE section. First the gas sampling probes will be sampled, then the shallow groundwater probe, then finally the observation well all on the same day. For the pesticide/nitrate locations the gas sampling probe will not provide pesticide and nitrate data but will give CFC data which will also be obtained from the shallow groundwater probe and observation well. Follow up lysymeter and sediment sampling would proceed later in FY98 at the pesticide/nitrate locations as discussed above in the UNSATURATED ZONE section.
Transport modelingUnsaturated-zone transport models provide a method to obtain the flux of chemicals through the unsaturated zone and across the water table, given concentration data and estimates of transport- model parameters. This provides a conceptual framework to assist in evaluating groundwater land-use data sets and to determine vulnerability of shallow groundwater as a function of unsaturated zone properties for unsampled land-use settings. At present, all the synthesis teams plan to use standard statistical and graphical techniques to summarize land-use groundwater data. The introduction of a modeling or deterministic viewpoint in analyzing these data will provide recommendations for grouping unsaturated zone parameters for additional statistical analysis.
In FY96 a screening model was developed to compare the movement of individual compounds (eg. MTBE compared to BTEX) for hypothetical unsaturated zone properties. This screening model allows the chemical to partition in aqueous, adsorbed, or gaseous phases and the chemical can be transported via advection in the aqueous phase and via diffusion in either the aqueous or gaseous phases. First order reactions (eg. biodegradation) involving the chemical can also be simulated. The chemical source is modeled by specifying the concentration at the land surface boundary. The chemical is assumed to move across the water-table boundary by advection. An analytical solution is available for the transient case assuming a homogeneous unsaturated zone and for the steady state case for a layered unsaturated zone. The solutions relate mass loading at land surface and mass loading at the water table as a function of unsaturated zone thickness, recharge rate, diffusion coefficient, phase partitioning properties, and microbial or chemical reactions. This model provides a start to interpret data from our probe nests and land use survey. Additional modeling and its relation to other NAWQA studies is discussed in the section Application to NAWQA land use surveys below.
In FY97 we will continue the comprehensive investigation of the Glassboro region 72 well land use-survey. The core collected from the unsaturated zone reveal stratigraphic layering and we plan to complete the laboratory analysis of the cores discussed above in the UNSATURATED ZONE - transport modeling parameters section. Unsaturated zone transport modeling will be used to analyze the large amount of data available from the cores and sampling network. The goal of this analysis is to provide a spatially dependent vulnerability index for the Glassboro region which relates land use, water quality, and unsaturated zone properties.
Training and Application to NAWQA land use surveysA new unsaturated zone training course is being developed by Lahvis, Baker, and Baehr as part of our Toxics Hydrology project and will be taught sometime in FY98. The course will focus on concepts, data collection, and modeling related to the transport of volatile organic compounds in the unsaturated zone. The original target audience for the course was scientists studying passive remediation at gasoline and other NAPL spill sites. liaison with synthesis teams and NAWQA Study Units (particularly select FY97-starting projects that will incorporate components of the comprehensive urban investigation) can easily be accomplished by expanding the scope of the course to include interpretation of NAWQA land use surveys.
In addition to the screening model discussed in this workplan, we have developed a more comprehensive numerical transport model as part of our Toxics Hydrology project that is applicable to this endeavor. Colleague review has recently been completed for:
Publication of the Open-file report is expected this year. R-UNSAT has capabilities to incorporate stratigraphic layers and heterogeniety with respect to various parameters which would be required to simulate a general field condition.
Sampling of streamsThe LINJ core project maintains an intensive fixed site in the Glassboro Region on the Upper Great Egg Harbor River at Sicklerville. The data is summarized in Table 3
TABLE 3-- SUMMARY OF WATER QUALITY DATA FROM INTENSIVE FIXED SITE ON THE UPPER GREAT EGG HARBOR RIVER AT SICKLERVILLE
VOC's-- 16 samples collected from 4/22/96 to 10/9/96
Continuation of the intensive fixed site at Sicklerville is proposed for FY97. The NJ District annual report has nutrient data for 2 other streams in the Glassboro region which indicate similar concentration levels for NO3. In addition the South Branch of Big Timber Creek at Blackwood Terrace was sampled for VOCs on April 2, 1996 and the results are as follows:
Chloromethane E0.04 (in ug/L)
It is anticipated that pesticide occurrence is maximum in early summer because of spring applications and VOC occurrence is maximum in mid winter because of cooler winter temperature. Because of the low VOC concentrations reported at Sicklerville, two streams in somewhat more urbanized portions of the Glassboro Region, Mantua Creek at Pitman and the South Fork of Big Timber Creek at Blackwood Terrace will be added to the LINJ-VOC stream synoptic (which includes Sicklerville) planned for late-January 1997. Five streams in the Glassboro region will be added to the LINJ pesticide/nutrient synoptic planned for early summer 1997. This synoptic will only be for SH 2001 pesticides because of the low concentrations of SH 2050 pesticides detected at Sicklerville. The five locations will be selected based on data from the extended land use survey and will be on the following Glassboro Region streams: Mantua Creek, South Branch of Big Timber Creek, Upper Great Egg Harbor River, Little Ease Run, Still Run, Chestnut Branch, Black Water, Four Mile Run, Scotland Run, and the Maurice River at Norma. Upon evaluating this synoptic data and the intensive fixed site time series, recommendations for additional water quality sampling, ecological surveys, and stream transport modeling will be made for FY98.
Transport modeling and surface/groundwater interactionThe study area (approximately 700 square miles) groundwater flow model has been applied to analyze the contributing areas for the major streams in the Glassboro Region. Conceptual flow paths originating from the water table and ending at a predicted stream location (or water supply well) can be defined. The time required for water to move along any portion of such a flow path can be estimated. This analysis is prerequisite to future models that would have finer resolution at a scale relevant to a particular stream.
Sampling stormwater and detention basinsDetention basins control runoff in new residential communities and commercial developments and are prominent in the Glassboro Region. Stormwater collected in the basins recharge the Cohansey Aquifer. The basins are excavated and therefore the residence time of collected water in the unsaturated zone is reduced. This in combination with the anticipation that VOC and pesticide concentrations in stormwater are significant give reason to evaluate detention basins as sources of contaminants to surficial aquifers.
In FY97 detention basin sampling will be initiated for the Glassboro region by beginning a synoptic network at approximately 6-10 basins. Residential area basins where lawn care is prevalent and commercial basins draining areas with extensive asphalt paving will be represented. Groundwater will be sampled beneath each basin near the center or on its downgradient side via geoprobe or drivepoint and analyzed for VOC, SH 2001 pesticides, nutrients, major ions, pH, alkalinity, and DO. The basins will be sampled once in FY97. The scope of this work for FY98 will be determined based on the FY97 study.
Information related to NAWQA can be obtained from:
NAWQA Project Chief, USGS